VCF to proteinDB

[husensofteng]

Convert a VEP annotated VCF into protein sequences.

Input:

• GTF file (matching the one used to annotated the VCF file)
• VCF file (VEP-annotated, e.g from ENSEMBL or gnomAD)
• Genome fasta file

Process:

• Convert the GTF file into fasta based on the given genome assembly.
• For each variant change the variant position in the transcript sequence and translate

Output:

• fasta file containing protein sequences per variant per affected transcript.

[husensofteng]

The main difference with COSMIC and cBioportal is:

• This is based on VCF files
• The annotations in the ENSEMBL VCF files only specify the affected transcript and the consequence type whereas the COSMIC files also specify the affected CDS position. Therefore, here the GTF file has to be used instead of the all_genes.fasta to re-calculate the affected CDS position.

[husensofteng]

Done Example command: python pypgatk_cli.py vcf-to-proteindb --config_file config/ensembl_config.yaml --vep_annotated_vcf testdata/test.vcf --input_fasta testdata/test.fa --gene_annotations_gtf testdata/test.gtf --output_proteindb testdata/proteindb_from_ENSEMBL_VCF.fa